Oracle's $6.7 Billion Bid for BEA Turned Down

andy1307 writes to tell us that according to the Mercury News, Oracle has made an unsolicited bid to buy BEA Systems for about $6.7 billion. BEA confirmed that it rejected the $17 a share bid as too low. "BEA told Phillips that its board of directors believes BEA 'is worth substantially more to Oracle, to others and, importantly, to our shareholders than the price indicated in your letter.' Oracle's aggressive bid may be an attempt to pre-empt an acquisition by others, Finley said. Those named in the past as potential suitors include IBM, the German software company SAP AG and Hewlett-Packard. Trip Chowdhry of Global Equity Research said he expects a counterbid from SAP, which he said needs BEA to survive. 'If they don't get BEA, probably in two years SAP will be on the block to sell itself,' Chowdhry predicted. Oracle needs to keep BEA out of competitors' hands, he said. Chowdhry said the offer currently 'is not right. Probably at $21 the deal will get done.'"

Re:Why?

[the eric conspiracy]

If you are selling enterprise products for a million dollars a pop that are based on J2EE technologies there are a lot of companies out there that don't want to hear that they are getting something based on FOSS.

This is the space Oracle plays in. By buying BEA they get a bunch more of this kind of customer.

To them you can bet it is worth 6.7 billion.

It was reported as a done deal

[DreadfulGrape]

Wow, did the business press ever jump the gun on this one. The headline in the Wall St. Journal this morning was "Oracle Buys BEA".

Why not Red Hat?

[Envy Life]

I'm not sure why Oracle wants BEA Systems, which appears to be antiquated and riddled with issues at this point. As other posters have indicated, switching to JBoss is not only successful it saves a ton of money. It was near a year ago Oracle bid for JBoss, losing out to Red Hat, then created their own "Unbreakable Linux" distribution based on Red Hat Linux. If they're willing to plunk down $6.8 billion for BEA Systems, Red Hat, at a market cap of $4.15 billion is not only a relative bargain but seems to be a no-brainer.

A Developers Perspective

[byronf]

I've been in the Java "Enterprise" domain for nearly ten years now. I first used BEA's weblogics back in 1999 (not even sure if it was owned by BEA yet). It was a good product, respected standards, and seemed to be development friendly. I considered it the best servlet engine at the time.

A couple of years ago I did a contract gig for a fortune 500 company that wanted to create corporate internal and external web portals. The company purchased the entire BEA middleware stack from Weblogics to Portal to Integration. The contract was several million dollars, with a support contract of several hundred thousand dollars a year. What a nightmare, we were working on weblogics 8.X using their workshop IDE (I called it Workswap), at the time you couldn't develop BEA Portal apps in any other IDE. The entire stack was buggy, and unreliable, including the IDE. However, workshop was pretty, and the BEA reps enjoyed demonstrating to the managers how simple it was to visually drop down a portlet, or a webservice, or whatever, and pow! Enterprise ready. The reality of course was much different. I decided then that BEA stopped being a developers company, and became a marketers company. Don't get me wrong, they are not the first, and will probably make bundles of money for a time, but it won't last.

These days I don't see any reason to purchase middle ware in the Java domain. You could make an argument for buying specialized tools or libraries, but not the big heavy applications servers with all the additional cruft that these companies make big money from. I see no advantage these products provide over what is freely available, well established, and standards based. The middle ware companies will preach support, and this strikes a chord with some managers, but it a complete fallacy. I've seen the valuable time of a company's senior developers wasted jumping through the hoops of a support process only to get some patch that works around a bug in the product. It would have taken much less time for the same developers to have been able to simple look at the source, and work the problem themselves. If BEA was smart they would sell now.

Get real

[einar2]

Yeah, and pigs can fly...

Sorry folks, we speak here about enterprise computing. Nobody really cares for open source until it is a proven product. And JBoss is not in this league. If you run several million transactions per day and each transaction makes you one EUR or more, there is nothing open source can give you:

• Nobody else with such a load is using JBoss.
• There is not enough support for your installation (every hour downtime is expensive ...)
• You cannot hire people off the street who are familiar with the product.
• Selling open source to your management is difficult. Typically, there is no company big enough behind an open source product to be legally responsible for it.
• Open source is no added value. If your run several million transactions a day then your are not a software company. Your competence lies elsewhere. Such companies never ever modify the source code of a product. Why should they? Modifying the code just makes the previously mentioned points worse...

BEA has a track record for enterprise products which in my personal experience is better than the one of IBM. IBM can get it right but it takes five years and several versions of their product. BEA can get introduce a new product in reasonable time and you do not have to wait three releases until it runs stable.